Eli Ben-Sasson

From theoretical computer science to cryptocurrency

Ben-Sasson earned his PhD in theoretical computer science from the Hebrew University of Jerusalem in 2001. He subsequently held research positions at Harvard, MIT and the Institute for Advanced Study in Princeton, and later became a professor of computer science at Technion.

His research concerned the relationship between computation and proof. Running a complicated calculation can require substantial work. Checking a proof that the calculation followed the rules can be much less demanding. Making that difference useful is central to his contribution to cryptography.

For blockchains, the appeal is practical. A network needs participants to agree that transactions are valid, but requiring everyone to repeat every operation limits how much activity it can handle. Proof systems offer another way to organise that verification.

The privacy question behind Zcash

Ben-Sasson was a co-inventor of the Zerocash protocol and a founding scientist of Zcash. That work explored a difficult problem for digital money: how to verify that a payment follows the rules while keeping sensitive transaction information private.

Zero-knowledge proofs make it possible to establish that a statement is true without revealing the private information used to prove it. In a payment system, this can help separate the need to check validity from the need to expose transaction details.

That distinction also helps explain the wider significance of his research. Privacy and public verification do not always have to pull in opposite directions. Carefully designed cryptography can support both, although the guarantees depend on the particular system and how it is used.

What STARK proofs changed

Ben-Sasson co-authored the 2018 STARK paper with Iddo Bentov, Yinon Horesh and Michael Riabzev. Their work described a practical approach to verifying computations efficiently without relying on a trusted setup.

The name stands for Scalable Transparent ARguments of Knowledge. “Scalable” concerns how the work grows as computations become larger. “Transparent” means the setup does not depend on a secret that participants must trust someone to handle correctly.

For a reader using a blockchain application, the mathematics can stay behind the scenes. The important idea is that substantial computation can be accompanied by a proof that is comparatively efficient to check. That creates opportunities to support more activity without asking every verifier to perform the original work.

Generating those proofs still takes resources. STARKs change the balance between doing work and checking it; they do not make computation free.

Turning research into StarkWare

In 2018, Ben-Sasson co-founded StarkWare with Uri Kolodny, Alessandro Chiesa and Michael Riabzev. The company gave the research an engineering setting: proof systems had to become software that developers could build with and applications could depend on.

Ben-Sasson now serves as StarkWare’s CEO. The company develops Cairo, STARK provers and the SHARP proving service, and contributes to Starknet and its developer tools.

These parts serve different purposes. Developers need ways to express the computations their applications perform. Provers produce evidence that those computations were executed correctly. The surrounding infrastructure makes that process usable beyond an individual research demonstration.

Moving between these layers is a significant part of StarkWare’s work. A promising proof system becomes useful to a wider audience only when the tools around it are reliable and accessible.

Starknet and the work of scaling Ethereum

Starknet brought STARK-based validity proofs into the Ethereum scaling ecosystem. Its development illustrates how proof research can support applications that execute activity outside Ethereum’s base layer and use proofs to verify the resulting state changes.

For application developers, the opportunity is to support more activity while reducing the amount of computation that must be repeated on the base layer. For users, the experience still depends on practical details such as fees, wallets, bridges and transaction reliability.

Those details matter when assessing any scaling system. A validity proof addresses whether a computation followed the rules. Questions about data availability, upgrades and transaction ordering require their own answers.

StarkWare and Starknet should also be distinguished. StarkWare is a company developing technology and contributing to the ecosystem. Starknet is the network around which developers, users and other contributors build. Ben-Sasson’s role is influential, but the work should not be credited to him alone.

Bitcoin and the next security questions

Ben-Sasson’s public work also examines how proof technology and cryptographic research could support Bitcoin. This includes discussion of scaling and preparations for future quantum-computing risks.

One recent example came on August 26, 2026, when StarkWare reported that a transaction using Avihu Levy’s Quantum-Safe Bitcoin construction had been mined on Bitcoin mainnet. The announcement credited Levy with the method and StarkWare engineer Tomer Giladi with work on its implementation.

The scope of that result matters. It demonstrated a particular construction under existing Bitcoin rules; it did not make the entire Bitcoin network quantum-safe. The method also did not use STARK proofs.

The distinction is useful when reading about Ben-Sasson’s work. Research proposals, working demonstrations and network-wide changes are different stages of progress. Each can be significant without implying that the next stage has already happened.

The thread connecting his work

Across privacy research, STARK proofs and StarkWare, the recurring question is how people can verify more while relying less on another party’s assurances. Different projects apply that question to different problems.

Some focus on keeping information private. Others aim to make computation easier to verify or prepare systems for changing security assumptions. Ben-Sasson’s career connects those research questions with the continuing work of turning them into usable software.